Researchers in South Korea have developed an artificial intelligence model to predict the virulence of the tomato yellow leaf curl virus (TYLCV) [1, 2].
The tool provides a way to anticipate how aggressive specific virus strains will be, which is critical for developing disease-management strategies to protect global tomato yields [1, 2].
Professor Balachandran Manavalan and his Computational Biology and Bioinformatics (CBBL) research team created the system, named DeepTYLCV [1, 2]. The team operates within the Department of Integrative Biotechnology at Sungkyunkwan University [1, 2].
DeepTYLCV is designed to be both accurate and interpretable, allowing scientists to understand the factors that contribute to the virulence of the virus [1, 2]. By predicting these outcomes, the model assists in the identification of high-risk strains before they cause widespread crop failure [1, 2].
This development addresses a significant challenge in agricultural science, where the rapid mutation of plant viruses often outpaces traditional laboratory testing [1, 2]. The AI-driven approach allows for faster screening of viral sequences to determine the potential impact on plant health [1, 2].
The research team focused on the TYLCV because of its ability to severely reduce tomato productivity in various climates [1, 2]. By utilizing the DeepTYLCV model, agriculturalists can better implement targeted interventions to mitigate the spread of the most virulent strains [1, 2].
“The tool provides a way to anticipate how aggressive specific virus strains will be.”
The creation of DeepTYLCV represents a shift toward predictive agriculture, where machine learning reduces the reliance on time-consuming physical trials. By identifying virulent strains digitally, researchers can accelerate the development of resistant tomato varieties and more effective pesticide or biological controls, potentially stabilizing food supplies in regions heavily reliant on tomato production.
